****README

* Replication dofile of the paper:
* Agricultural Transformation and Farmers' Expectations:  Experimental Evidence from Uganda 
* Jacopo Bonan, Harounan Kazianga, Mariapia Mendola
* July 2025

The dataset included in "dataset_crop_panel_may25.dta" is a cross-section. The unit of observation is a oilseed-farmer. Hence, each farmer is observed  four times, one for each oilseed. Oilseeds are identified by "oilseed_id": soybean (oilssed_id==1), sunflower (oilssed_id==2), sesame (oilssed_id==3), grouondnut (oilssed_id==4)

The farmer ID is "id", the unit of randomization is the subcounty ("sub_county_code_n"), strata are districts ("distr_code"). The treatment variable is "treatment_status".

Variables ending in "base" are measured at the baseline. Variables not ending in "base" are measured at the endline.

Variables ending in "_" are oilseed-specific, i.e., will take different values for the four oilseeds.  
Variables not ending in "_" are farmer-specific.

Variables are labelled as in the paper. 
 

